def test_eval_parametric(self):
data = np.array([1, 1, 1, 1, 1, 1, 1], dtype=np.float32).reshape(
(1, 7))
spn = (Gaussian(mean=1.0, stdev=1.0, scope=[0]) *
Exponential(l=1.0, scope=[1]) *
Gamma(alpha=1.0, beta=1.0, scope=[2]) *
LogNormal(mean=1.0, stdev=1.0, scope=[3]) *
Poisson(mean=1.0, scope=[4]) * Bernoulli(p=0.6, scope=[5]) *
Categorical(p=[0.1, 0.2, 0.7], scope=[6]))
ll = log_likelihood(spn, data)
tf_ll = eval_tf(spn, data)
self.assertTrue(np.all(np.isclose(ll, tf_ll)))
spn_copy = Copy(spn)
tf_graph, data_placeholder, variable_dict = spn_to_tf_graph(
spn_copy, data, 1)
with tf.Session() as sess:
sess.run(tf.global_variables_initializer())
tf_graph_to_spn(variable_dict)
str_val = spn_to_str_equation(spn)
str_val2 = spn_to_str_equation(spn_copy)
self.assertEqual(str_val, str_val2)
def tensorflow():
import numpy as np
spn = create_SPN()
test_data = np.array([1.0, 0.0, 1.0]).reshape(-1, 3)
from spn.gpu.TensorFlow import eval_tf
lltf = eval_tf(spn, test_data)
print("tensorflow ll", lltf, np.exp(lltf))
from spn.algorithms.Inference import log_likelihood
from spn.gpu.TensorFlow import optimize_tf
optimized_spn = optimize_tf(spn, test_data)
lloptimized = log_likelihood(optimized_spn, test_data)
print("tensorflow optimized ll", lloptimized, np.exp(lloptimized))
def test_eval_gaussian(self):
np.random.seed(17)
data = np.random.normal(10, 0.01,
size=2000).tolist() + np.random.normal(
30, 10, size=2000).tolist()
data = np.array(data).reshape((-1, 10))
data = data.astype(np.float32)
ds_context = Context(meta_types=[MetaType.REAL] * data.shape[1],
parametric_types=[Gaussian] * data.shape[1])
spn = learn_parametric(data, ds_context)
ll = log_likelihood(spn, data)
tf_ll = eval_tf(spn, data)
self.assertTrue(np.all(np.isclose(ll, tf_ll)))
def test_eval_histogram(self):
np.random.seed(17)
data = np.random.normal(10, 0.01,
size=2000).tolist() + np.random.normal(
30, 10, size=2000).tolist()
data = np.array(data).reshape((-1, 10))
data[data < 0] = 0
data = data.astype(int)
ds_context = Context(meta_types=[MetaType.DISCRETE] * data.shape[1])
ds_context.add_domains(data)
spn = learn_mspn(data, ds_context)
ll = log_likelihood(spn, data)
tf_ll = eval_tf(spn, data)
self.assertTrue(np.all(np.isclose(ll, tf_ll)))
memory = Memory(cachedir="cache", verbose=0, compress=9)
@memory.cache
def learn(data, ds_context):
spn = learn_structure(data, ds_context, get_split_rows_KMeans(), get_split_cols_RDC(), create_histogram_leaf)
return spn
if __name__ == '__main__':
np.random.seed(17)
data = np.random.normal(10, 0.01, size=2000).tolist() + np.random.normal(30, 10, size=2000).tolist()
data = np.array(data).reshape((-1, 10))
ds_context = Context(meta_types=[MetaType.REAL] * data.shape[1])
ds_context.add_domains(data)
spn = learn(data, ds_context)
py_ll = likelihood(spn, data, histogram_likelihood)
tf_out, time = eval_tf(spn, data, log_space=False, save_graph_path='tfgraph', trace=True)
tf_out, time = eval_tf(spn, data, log_space=False, save_graph_path='tfgraph', trace=True)
tf_out_log, time2 = eval_tf(spn, data, log_space=True, save_graph_path='tfgraph')
print("results are similar for TF and Python?", np.all(np.isclose(py_ll - np.log(tf_out), 0)))
print("time in ns", time)
print("results are similar for Log TF and Python?", np.all(np.isclose(py_ll - tf_out_log, 0)))
if __name__ == '__main__':
add_histogram_inference_support()
np.random.seed(17)
data = np.random.normal(10, 0.01, size=2000).tolist() + np.random.normal(
30, 10, size=2000).tolist()
data = np.array(data).reshape((-1, 10))
data[data < 0] = 0
data = (data * 1).astype(int)
ds_context = Context(meta_types=[MetaType.DISCRETE] * data.shape[1])
ds_context.add_domains(data)
data[:, 0] = 0
data[:, 1] = 1
spn = learn(data, ds_context)
spn = create_histogram_leaf(data[:, 0].reshape((-1, 1)), ds_context, [0], alpha=False, hist_source="kde") * \
create_histogram_leaf(data[:, 1].reshape((-1, 1)), ds_context, [1], alpha=False, hist_source="kde")
spn = 0.3 * create_histogram_leaf(data[:, 0].reshape((-1, 1)), ds_context, [0], alpha=False, hist_source="kde") + \
0.7 * create_histogram_leaf(data[:, 0].reshape((-1, 1)), ds_context, [0], alpha=False, hist_source="kde")
py_ll = log_likelihood(spn, data)
tf_graph, placeholder = spn_to_tf_graph(spn, data)
log_tf_out = eval_tf(tf_graph, placeholder, data)
print("results are similar for Log TF and Python?",
np.all(np.isclose(py_ll, log_tf_out)))
